resolve-session
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data retrieved from the DaVinci Resolve API, including project names, timeline metadata, and media pool clip titles. It also instructs the agent to follow instructions contained within the
remediationfield of connection error messages. This represents a potential surface for indirect prompt injection if an attacker were to maliciously name projects or clips to influence agent behavior. - Ingestion points: Data retrieved via
resolve_control,project_manager, andtimelinetools (SKILL.md). - Boundary markers: None identified in the prompt instructions.
- Capability inventory: Resolve project management, timeline editing, and media pool manipulation.
- Sanitization: No explicit sanitization or filtering of Resolve metadata is described.
Audit Metadata